practice, and that we can with considerable propriety apply to him the words of the learned friend of Tyrwhitt, (see Brunck's note on Sophocles' Electr. 21.) whose critical acumen is not surpassed, perhaps not equalled, by any scholar of the present day; and whose learning and judgment were, we will venture to say, little inferior to the learning and judgment of the author of the Preface to the "Musæ Cantabrigienses.". "Argumenta sumit (says he, speaking of a man somewhat hypercritical) ex poetarum depravatis versibus." Let us see how far the truth of this extends to the case before us in the first place, the only line (if we may judge from Sappho's fragments) that can, allowing it to be correct, support this objection, (and from his reading aye a little after with the Æolic digamma prefixed, he can have had no other in his eye,) is the following, as it is commonly read: ἀλλὰ καμμὲν γλῶσσα ἔαγε, λέπτον δ' Or, as he prefers; ἀλλὰ καμμὲν γλῶσσα Γἔαγε, λέπτον δ Did he believe this reading right? and that nothing was wanting to perfect the line, but the elegant insertion of the Æolic digamma? Toup, 'tis true, thought the passage, in the state in which he received it, corrupt,-he altered, and succeeded little better than our Editor, reading with Manutius, ἀλλὰ καμμὲν γλῶσσ ̓ ἐάγη, ἂν δὲ λέπτον The elision thus formed in the middle of the line, is intolerably harsh, and sanctioned (if we mistake not) by no legal authority; but, whatever Toup's blunder might be, he did not believe that a short vowel, either in a polysyllable, or a monosyllable, could be elided at the end of any line, but the third in the Sapphic stanza. The oldest editions of Sappho's fragments read the line thus: ἀλλὰ καμμὲν γλῶσσ ̓ ἔαγ', ἂν δὲ λέπτον Then the whole will run thus: ἀλλὰ καμμὲν γλῶσσ ̓ ἔαγ', ἂν δὲ λέπτον αὐτίκα χρῶ πᾶς ὑποδεδςόμακεν. VOL. IV. No. vII. F "Sed et lingua fracta est; et statim subtilis ignis sensim subiit cutem." More freely, and giving the full sense of the passage: "Sed et lingua mea torpet debilitata, atque illicò subtilis ignis sensim, ut ita dicam, et pedetentim corpus meum subiit, et totam me surripuit." The force of the original is beautifully preserved in Catullus's translation; "Lingua sed torpet; tenuis sub artus There is not a word in the Latin language, which could have corresponded with úrodedgóμaxev so closely, so exactly, as "dimanat ;" the component parts of the two words perfectly coincide; "dimano" strictly is "to flow or gush gradually;" many other words so compounded may be adduced, which imply the idea of sensim; thus we have, " direpo," to creep gradually, slowly; so also "dilapsus," as in Livy, " Ædem vetustate dilapsam refecit:" where delapsam would have been improper: "dimoveo," too, as in Horace about Regulus, "Non aliter tamen Dimovit obstantes propinquos Et populum reditus morantem; Quam si clientum longa negotia, Dijudicatâ lite, relinqueret, Tendens Venafranos in agros Aut Lacedæmonium Tarentum." We have quoted the whole passage, in order that it may more immediately appear from the connexion the word has with every part of it, that it must necessarily signify « gradually removed;" if it were not so, the whole would be absurd; otherwise how could it be said of Regulus, "Dimovitpropinquos-non aliter quàm si relinqueret, &c.-tendens, &c." the context imperiously calls for this signification. To return to the matter in hand, evidence is strongly in our favor, that Catullus read the line as the oldest editions, and amongst them those of H. Stephens did. But why was not Toup content with Stephens's reading and what could induce our author to reject every previous reading? Toup thought that the penult of says 'was uniformly long, for which he substituted" prorsùs άuérgws,” (as Brunck observes,) éáyn; which word, by the way, occurs in Homer with the penult both long and short: νωθὴς, ᾧ δὴ πολλὰ περὶ ῥόπαλ ̓ ἀμφὶς ἐάγη. Iliad 4. 558. νῦν δέ μοι ἐν χείρεσσ ̓ ἐάγη ξίφος· ἐκ δέ μοι ἔγχος. r. 367. and it was not, as the Editor supposes, on account of the hiatus, that Toup rejected eye. But why any alteration from the original aye? we know that the form ya, contracted from aya, exists: if so, the quantity of the a in the uncontracted word must be short; for every school-boy, who has been drilled through the clumsiness. of modern Greek Grammar, will tell you," si vocalis longa, aut diphthongus, sequitur e, fit contractio tollendo ." Consequently, if the penult of aya was long, the contraction would be aya, which (except in the Doric dialect) we do not believe to exist; but, since the contracted form is ya and not åya, the penult of aya must be short. Unluckily for Toup, our Editor, and his digamma, we find in the Cyclops of Euripides the very word, of which Brunck had said, “ Tmesis est: κατέαγε μὲν γλῶσσα.” κακόν γε πρὸς κακῷ, τὸ κράνιον Παίσας κατέαγα. Cycl 678. Here, it must be confessed, the passage before us does not positively determine, whether xaréaya is used in a passive or an active sense; either way is defensible; but this is of no consequence to us in the present case, since the quantity is all we are contending for, which, we believe, we have satisfactorily decided. Those who are inclined, either from ignorance, or obstinacy, to defend the final elision, must first prove, that the penult of aye is long, uniformly long. "Tis true, that by the "hocuspocus" of transposition, the authority which I have produced might apparently be rendered ineffectual, viz. by reading, κακόν γε πρὸς κακῷ, τὸ κράνιον κατέαγα παίσας But an alteration like this, except upon the authority of MSS. would be impudent and absurd. And when we consider that the penult of the word must be short, if the canon about contraction before quoted be true, (of which there can be little doubt) we are strongly inclined to believe, that the passage in the Cyclops is correct. The MSS. from which Aldus printed his edition, evidently read it thus. Besides, there is another very strong reason why the line ought not to be read, ἀλλὰ καμμὲν γλῶσσα ἔαγε, λέπτον δ'. By this alteration, granting for a while that the final elision is allowable, there is a word totally expunged, which is of high importance in its place, and which Catullus (if we may judge from his translation) read in his copy, ἀλλὰ καμμὲν γλῶσσ ̓ ἔαγ. ̓ΑΝ δὲ λέπτον is the reading of Stephens, and of the earliest editions: Stephens, it appears, is the only man who knew any thing about the matter, as he has clearly shown by his accentuation of 'AN; Toup, we fear, (" pace tanti viri dicatur") passed over the word in silence, not knowing what to make of it; according to his accentuation, *ΑΝ δὲ λέπτον αὐτίκα χρῶ πῦς ὑποδεδρό μακεν. We should translate," atque illicò subtilis ignis sensim cutem meam subiisset;" which was evidently not Sappho's meaning. The av, which is to be referred to xg, is a fragment of the preposition avà, and often stands before an initial consonant : thus in Homer: I Βῆρ ̓ ἴμεν *ΑΝ· τε μάχην καὶ ̓ΑΝΑ κλόνον ἐγχειάων. Θεοδότων ἔργων κέλευθον ̓ΑΝ καθαράν. Isthm. v. 28. I It may be needless to say, that 'AN, in this case, receives its accent from the enclitic Ti. In the passage before us, 'AN signifies UP and THROUGHOUT, viz. FROM TOP TO BOTTOM; so in the beginning of the Iliad; Where we should render 'ANA in English by right through, viz. from top to bottom, so as to leave no part untouched: 'AN Xp therefore signifies, right through my body, i. e. so that every part was affected. On the supposition that this difficulty is removed, there remains no objection on the score of any other line in the fragments of Sappho, so that, with considerable security we may lay down the following canon; "In Greek Sapphic verse, no final elision, either of a monosyllable or polysyllable is allowed, except in the third line." In the Ode of Erinne, (who was cotemporary with Sappho,) there is not the slightest shadow of an instance; and from the general strain of Greek poetry, we know that the Greeks were much more chary of final elisions than the Latins; the latter admit them even in Heroics, whereas (if we mistake not) the former have not a single instance, where either a monosyllable or a polysyllable is elided at the end of an hexameter line. As to the hiatus, of which our Editor complains so much ; "Vocalium hiatus nimis licenter quidam admiserunt; quod in constrictis hujusmodi metris minùs rectè fieri judicamus ;”—we certainly agree with him, as far as single vowels are concerned; but our opinion is, that where a final diphthong precedes an initial vowel, it is uniformly short, and that instances of this kind ought to be avoided no more in Greek Sapphics than in Greek Heroics. Thus, in the fragment preserved by Longinus, we have palvoμAľ "Añvous; and in an extract from Sappho, found in Macrobius's Saturnal. book v. §. 21. |